Religion As a Cultural System
by Clifford Geertz
Any attempt to speak without speaking any particular language is not more hopeless than the attempt to have a religion that shall be no religion in particular . . . . Thus every living and healthy religion has a marked idiosyncrasy. Its power consists in its special and surprising message and in the bias which that revelation gives to life. The vistas it opens and the mysteries it propounds are another world to live in; and another world to live in--whether we expect ever to pass wholly over into it or no--is what we mean by having a religion.
SANTAYANA, Reason in Religion
I
Two characteristics of anthropological work on religion accomplished since the second world war strike me as curious when such work is placed against that carried out just before and just after the first. One is that it has made no theoretical advances of major importance. It is living off the conceptual capital of its ancestors, adding very little, save a certain empirical enrichment, to it. The second is that it draws what concepts it does use from a very narrowly defined intellectual tradition. There is Durkheim, Weber, Freud, or Malinowski, and in any particular work the approach of one or two of these transcendent figures is followed, with but a few marginal corrections necessitated by the natural tendency to excess of seminal minds or by the expanded body of reliable descriptive data. But virtually no one even thinks of looking elsewhere--to philosophy, history, law, literature, or the "harder" sciences--as these men themselves looked, for analytical ideas. And it occurs to me, also, that these two curious characteristics are not unrelated. If the anthropological study of religion is in fact in a state of general stagnation, I doubt that it will be set going again by producing more minor variations on classical theoretical themes. Yet one more meticulous case in point for such well-established propositions as that ancestor worship supports the jural authority of elders, that initiation rites are means for the establishment of sexual identity and adult status, that ritual groupings reflect political oppositions, or that myths provide charters for social institutions and rationalizations of social privilege, may well finally convince a great many people, both inside the profession and out, that anthropologists are, like theologians, firmly dedicated to proving the indubitable. In art, this solemn reduplication of the achievements of accepted masters is called academicism; and I think this is the proper name for our malady also. Only if we abandon, in a phrase of Leo Steinberg's, that sweet sense of accomplishment which comes from parading habitual skills and address ourselves to problems sufficiently unclarified as to make discovery possible, can we hope to achieve work which will not just reincarnate that of the great men of the first quarter of this century, but match it.1 The way to do this is not to abandon the established traditions of social anthropology in this field, but to widen them. At least four of the contributions of the men who, as I say, dominate our thought to the point of parochializing it--Durkheim's discussion of the nature of the sacred, Weber's Verstehenden methodology, Freud's parallel between personal rituals and collective ones, and Malinowski's exploration of the distinction between religion and common sense--seem to me inevitable starting-points for any useful anthropological theory of religion. But they are starting-points only. To move beyond them we must place them in a much broader context of contemporary thought than they, in and of themselves, encompass. The dangers of such a procedure are obvious: arbitrary eclecticism, superficial theory-mongering, and sheer intellectual confusion. But I, at least, can see no other road of escape from what, referring to anthropology more generally, Janowitz has called the dead hand of competence.2
In working toward such an expansion of the conceptual envelope in which our studies take place, one can, of course, move in a great number of directions; and perhaps the most important initial problem is to avoid setting out, like Stephen Leacock's mounted policeman, in all of them at once. For my part, I shall confine my effort to developing what, following Parsons and Shils, I refer to as the cultural dimension of religious analysis.3 The term "culture" has by now acquired a certain aura of ill-repute in social anthropological circles because of the multiplicity of its referents and the studied vagueness with which it has all too often been invoked. (Though why it should suffer more for these reasons than "social structure" or "personality" is something I do not entirely understand.) In any case, the culture concept to which I adhere has neither multiple referents nor, so far as I can see, any unusual ambiguity: it denotes an historically transmitted pattern of meanings embodied in symbols, a system of inherited conceptions expressed in symbolic forms by means of which men communicate, perpetuate, and develop their knowledge about and attitudes toward life. Of course, terms such as "meaning," "symbol," and "conception" cry out for explication. But that is precisely where the widening, the broadening, and the expanding come in. If Langer is right that "the concept of meaning, in all its varieties, is the dominant philosophical concept of our time," that "sign, symbol, denotation, signification, communication . . . are our [intellectual] stock in trade," it is perhaps time that social anthropology, and particularly that part of it concerned with the study of religion, became aware of the fact.4
II
As we are to deal with meaning, let us begin with a paradigm: viz., that sacred symbols function to synthesize a people's ethos--the tone, character, and quality of their life, its moral and aesthetic style and mood --and their world view--the picture they have of the way things in sheer actuality are, their most comprehensive ideas of order. In religious belief and practice a group's ethos is rendered intellectually reasonable by being shown to represent a way of life ideally adapted to the actual state of affairs the world view describes, while the world view is rendered emotionally convincing by being presented as an image of an actual state of affairs peculiarly well-arranged to accommodate such a way of life. This confrontation and mutual confirmation has two fundamental effects. On the one hand, it objectivizes moral and aesthetic preferences by depicting them as the imposed conditions of life implicit in a world with a particular structure, as mere common sense given the unalterable shape of reality. On the other, it supports these received beliefs about the world's body by invoking deeply felt moral and aesthetic sentiments as experiential evidence for their truth. Religious symbols formulate a basic congruence between a particular style of life and a specific (if, most often, implicit) metaphysic, and in so doing sustain each with the borrowed authority of the other.
Phrasing aside, this much may perhaps be granted. The notion that religion tunes human actions to an envisaged cosmic order and projects images of cosmic order onto the plane of human experience is hardly novel. But it is hardly investigated either, so that we have very little idea of how, in empirical terms, this particular miracle is accomplished. We just know that it is done, annually, weekly, daily, for some people almost hourly; and we have an enormous ethnographic literature to demonstrate it. But the theoretical framework which would enable us to provide an analytic account of it, an account of the sort we can provide for lineage segmentation, political succession, labor exchange, or the socialization of the child, does not exist.
Let us, therefore, reduce our paradigm to a definition, for, although it is notorious that definitions establish nothing, in themselves they do, if they are carefully enough constructed, provide a useful orientation, or reorientation, of thought, such that an extended unpacking of them can be an effective way of developing and controlling a novel line of inquiry. They have the useful virtue of explicitness: they commit themselves in a way discursive prose, which, in this field especially, is always liable to substitute rhetoric for argument, does not. Without further ado, then, a religion is:
(1) a system of symbols which acts to (2) establish powerful, pervasive, and long-lasting moods and motivations in men by (3) formulating conceptions of a general order of existence and (4) clothing these conceptions with such an aura of factuality that (5) the moods and motivations seem uniquely realistic.
(1) a system of symbols which acts to . . .
Such a tremendous weight is being put on the term "symbol" here that our first move must be to decide with some precision what we are going to mean by it. This is no easy task, for, rather like "culture," "symbol" has been used to refer to a great variety of things, often a number of them at the same time.
In some hands it is used for anything which signifies something else to someone: dark clouds are the symbolic precursors of an on-coming rain. In others it is used only for explicitly conventional signs of one sort or another: a red flag is a symbol of danger, a white of surrender. In others it is confined to something which expresses in an oblique and figurative manner that which cannot be stated in a direct and literal one, so that there are symbols in poetry but not in science, and symbolic logic is misnamed. In yet others, however, it is used for any object, act, event, quality, or relation which serves as a vehicle for a conception-the conception is the symbol's "meaning"--and that is the approach I shall follow here.5The number 6, written, imagined, laid out as a row of stones, or even punched into the program tapes of a computer, is a symbol. But so also is the Cross, talked about, visualized, shaped worriedly in air or fondly fingered at the neck, the expanse of painted canvas called "Guernica" or the bit of painted stone called a churinga, the word "reality," or even the morpheme "-ing." They are all symbols, or at least symbolic elements, because they are tangible formulations of notions, abstractions from experience fixed in perceptible forms, concrete embodiments of ideas, attitudes, judgments, longings, or beliefs. To undertake the study of cultural activity--activity in which symbolism forms the positive content--is thus not to abandon social analysis for a Platonic cave of shadows, to enter into a mentalistic world of introspective psychology or, worse, speculative philosophy, and wander there forever in a haze of "Cognitions," "Affections," "Conations," and other elusive entities. Cultural acts, the construction, apprehension, and utilization of symbolic forms, are social events like any other; they are as public as marriage and as observable as agriculture.
They are not, however, exactly the same thing; or, more precisely, the symbolic dimension of social events is, like the psychological, itself theoretically abstractable from those events as empirical totalities. There is still, to paraphrase a remark of Kenneth Burke's, a difference between building a house and drawing up a plan for building a house, and reading a poem about having children by marriage is not quite the same thing as having children by marriage.6Even though the building of the house may proceed under the guidance of the plan or--a less likely occurrence--the having of children may be motivated by a reading of the poem, there is something to be said for not confusing our traffic with symbols with our traffic with objects or human beings, for these latter are not in themselves symbols, however often they may function as such.7No matter how deeply interfused the cultural, the social, and the psychological may be in the everyday life of houses, farms, poems, and marriages, it is useful to distinguish them in analysis, and, so doing, to isolate the generic traits of each against the normalized background of the other two.
So far as culture patterns, that is, systems or complexes of symbols, are concerned, the generic trait which is of first importance for us here is that they are extrinsic sources of information. By "extrinsic," I mean only that--unlike genes, for example--they lie outside the boundaries of the individual organism as such in that intersubjective world of common understandings into which all human individuals are born, in which they pursue their separate careers, and which they leave persisting behind them after they die. By "sources of information," I mean only that--like genes--they provide a blueprint or template in terms of which processes external to themselves can be given a definite form. As the order of bases in a strand of DNA forms a coded program, a set of instructions, or a recipe, for the synthesis of the structurally complex proteins which shape organic functioning, so culture patterns provide such programs for the institution of the social and psychological processes which shape public behavior. Though the sort of information and the mode of its transmission are vastly different in the two cases, this comparison of gene and symbol is more than a strained analogy of the familiar "social heredity" sort. It is actually a substantial relationship, for it is precisely because of the fact that genetically programmed processes are so highly generalized in men, as compared with lower animals, that culturally programmed ones are so important; only because human behavior is so loosely determined by intrinsic sources of information that extrinsic sources are so vital. To build a dam a beaver needs only an appropriate site and the proper materials--his mode of procedure is shaped by his physiology. But man, whose genes are silent on the building trades, needs also a conception of what it is to build a dam, a conception he can get only from some symbolic source--a blueprint, a textbook, or a string of speech by someone who already knows how dams are built--or, of course, from manipulating graphic or linguistic elements in such a way as to attain for himself a conception of what dams are and how they are built.
This point is sometimes put in the form of an argument that cultural patterns are "models," that they are sets of symbols whose relations to one another "model" relations among entities, processes or what-haveyou in physical, organic, social, or psychological systems by "paralleling," "imitating," or "simulating" them.8 The term "model" has, however, two senses--an "of " sense and a "for" sense--and though these are but aspects of the same basic concept they are very much worth distinguishing for analytic purposes. In the first, what is stressed is the manipulation of symbol structures so as to bring them, more or less closely, into parallel with the pre-established nonsymbolic system, as when we grasp how dams work by developing a theory of hydraulics or constructing a flow chart. The theory or chart models physical relationships in such a way--that is, by expressing their structure in synoptic form--as to render them apprehensible; it is a model of "reality." In the second, what is stressed is the manipulation of the nonsymbolic systems in terms of the relationships expressed in the symbolic, as when we construct a dam according to the specifications implied in an hydraulic theory or the conclusions drawn from a flow chart. Here, the theory is a model under whose guidance physical relationships are organized: it is a model for "reality." For psychological and social systems, and for cultural models that we would not ordinarily refer to as "theories," but rather as "doctrines," "melodies," or "rites," the case is in no way different. Unlike genes, and other nonsymbolic information sources, which are only models for, not models of, culture patterns have an intrinsic double aspect: they give meaning, that is, objective conceptual form, to social and psychological reality both by shaping themselves to it and by shaping it to themselves.
It is, in fact, this double aspect which sets true symbols off from other sorts of significative forms. Models for are found, as the gene example suggests, through the whole order of nature; for wherever there is a communication of pattern, such programs are, in simple logic, required. Among animals, imprint learning is perhaps the most striking example, because what such learning involves is the automatic presentation of an appropriate sequence of behavior by a model animal in the presence of a learning animal which serves, equally automatically, to call out and stabilize a certain set of responses genetically built into the learning animal.9 The communicative dance of two bees, one of which has found nectar and the other of which seeks it, is another, somewhat different, more complexly coded, example.10Craik has even suggested that the thin trickle of water which first finds its way down from a mountain spring to the sea and smooths a little channel for the greater volume of water that follows after it plays a sort of model for function.11 But models of--linguistic, graphic, mechanical, natural, etc., processes which function not to provide sources of information in terms of which other processes can be patterned, but to represent those patterned processes as such, to express their structure in an alternative medium--are much rarer and may perhaps be confined, among living animals, to man. The perception of the structural congruence between one set of processes, activities, relations, entities, and so on, and another set for which it acts as a program, so that the program can be taken as a representation, or conception--a symbol--of the programmed, is the essence of human thought. The intertransposability of models for and models of which symbolic formulation makes possible is the distinctive characteristic of our mentality.